- Email: [email protected]
A compact Compton suppression spectrometer using bismuth germanate scintillator
Nuclear Instruments and Methods 212 (1983) 241-244 North-Holland Publishing Company
241
A COMPACT C O M P T O N S U P P R E S S I O N S P E C T R O M E T E R U S I N G B I S M U T H GERMANATE SCINTILLATOR Mona MOHSEN Department of Physics, Section of Girls, King Abd EI Aziz University, Jeddah, P.O. Box 9030, Saudi Arabia
Received 31 March 1982 and in revised from 21 December 1982
Geometrical and random sampling methods of calculation have been used to estimate the dimensions of a Compton suppression spectrometer using bismuth germanate (BGO) scintillator. Similar calculations have been done using sodium iodide. It has been found that the dimensions of the BGO shield can be decreased to nearly half those of NaI(TI).
1. Introduction In -/-ray spectroscopy dealing with reactions or decay, the complexity of the spectra and the low yield of many transitions is difficult to analyze without increasing the peak-to-background ratio. One of the ways to enhance weak transitions with energies up to 4 MeV is in the suppression of the Compton continuum. Camp [1] designed a symmetrical Ge(Li)-NaI(T1) Compton suppression spectrometer, which consists of a cylindrical scintillator shield with a coaxial hollow core for a central detector at one side and the incoming y-ray beam at the opposite side. Such a set-up suffers from the presence of collinear holes which leads to a low suppression of double escape peaks as well as Compton scattering in the low and high energy parts of the spectrum. This disadvantage could be avoided in an asymmetrical geometry, where the central detector is oriented at right angles to the direction of the incoming radiation. Using this geometry, Ge-NaI(T1) Compton suppression spectrometers have been successfully constructed and their performance has been studied in radioactive decay [2], 7 - ' / c o i n c i d e n c e [3] and in experiments dealing with in-beam '/-ray spectroscopy at the external beam of a synchrocyclotron [4]. However, it will be of value especially for multiparameter measurements to reduce the size of such a spectrometer without affecting its function and this could be achieved by replacing NaI(T1) by bismuth germanate (BiaGe3012 known as BGO), which has recently proved to be a more efficient scintillator
[5]. Table 1 shows a comparison of the optical and mechanical properties of NaI(T1) and B G O scintillators. Although the energy and time resolution of NaI is better than BGO, yet the latter is characterized by high atomic number and density, which leads to an increase 0167-5087/83/0000-0000/$03.00 © 1983 North-Holland
Table 1 Comparison of NaI(T1) and BGO scintillators. Nal(TI) Energy resolution at 137Cs-photopeak (%) 7 Light output (relative) 100 Wavelength of maximum scintillation intensity(rim) 420 Refractive index at maximum intensity 1.85 Radiation length (cm) > 1.05 Decay constant (ns) 250 After glow (relative) 100 Density (g/cm 3) 3.67 Hygroscopicity hygroscopic
BGO 15 8-12 480 2.1 1.05 300 1 7.13 non-hygroscopic
in the stopping power and the photopeak efficiency of the BGO shield. Moreover, its full energy peak efficiency at 1.33 MeV has been measured by Evans [6] and found to be greater than that of NaI(TI) by a factor of 4.5. It should also be pointed out that B G O is free of problems associated with nonuniform dopant distribution. Weber et al. [7] identified the light emitted from B G O crystals with the 3P 1 ~ 1 S 0 electronic transition within the Bi3++ ions, giving rise to a " p u r e " scintillator. Moreover, Nitsche [8], Bortfeld et al. [9] and Nestor et al. [10] measured the light transmission in Bi4Ge3Ot2 for 0.5 mm, 0.65 mm and 17,5 mm sample thicknesses respectively as functions of light wavelength. Fig. 1 shows the light transmission in BGO taken from these references as a function of sample thickness at 480 nm, which represents the wavelength of maximum scintillation intensity [10]. Even for a sample thickness of 100 cm a light transmission greater than 65% can be obtained. Accordingly, these properties can then nominate
M. Mohsen / Compact Compton suppression spectrometer
242
I
-/.
i
80
i
Z
o_ 70
(n :E 01 Z
.< rr I-
.J
60
50
0 m
5
0.I
2
I
5
I0
2
Leng'rh
I00
5 fftnl
I000
Fig. 1. BGO light transmission as a function of sample length at 480 nm wavelength.
B G O for construction of a compact Compton suppression spectrometer, whose function depends on efficiency rather than on energy and time resolution. The aim of the present work is to present a comparative study for an asymmetrical Compton suppression spectrometer using G e - B G O as well as Ge-NaI(T1). For this purpose geometrical and random sampling methods have been applied to calculate the dimensions of each spectrometer as a function of its relative absorption.
2. Principle A Compton suppression spectrometer consists of a central Ge detector surrounded by a scintillation shield (BGO or NaI). G a m m a rays from a source outside are collimated before striking the Ge detector positioned at right angle to the direction of the beam. Compton scattered gamma rays from that detector can then interact with the surrounding scintillating material. If the resulting signal is above the noise level, then it can be used to reject the recording of any coincident signal in the Ge detector. Accordingly, full energy events only should in principle remain in the spectrum of the Ge detector. However, the Compton continuum is not completely eliminated since backscattered gamma rays can escape from the scintillation crystal through the entrance hole of the incident beam. Such gamma rays leave most of their energy in the Ge detector and this contribution appears at the Compton edge of the spectrum.
3. Geometrical method of calculation In order to estimate the dimensions of a Compton suppression spectrometer geometrically, gamma events in the germanium due to single Compton collisions are assumed. The continuum spectrum without shielding can be represented as do dE
do dO dR d E '
(1)
where o is the cross section for Compton collision, E is the kinetic energy of the recoil electron in the Ge detector and R is the angle of the scattered gamma relative to the incident gamma. The quantity d o / d R is the cross section for photons scattered between two cones at 8 and R + dR. The quantity d R ~ d E is obtained from E=h%[l-(l+c~(1-cosO)}
1],
(2)
where h u0 is the energy of incident gamma rays and c~ = h 1,o/mc 2.
To determine the expected effect of the anticoincidence shield in suppressing the continuum, the absorption of the scintillator around the Ge detector is investigated according to the scattering angle 0. The fractional absorption A / A o of the scattered gammas in the appropriate path length x ( 0 ) of the scintillator is calculated from A / A o = 1 - e -~x(°),
where # is the total linear absorption coefficient of the scintillator for the gamma energy ( h % - E) at angle 0.
M. Mohsen / Compact Compton suppression spectrometer
4. Random sampling of path length
direction of the g a m m a ray incident through a hole d i a m e t e r of 1 cm at a distance t cm from the detector. A s s u m i n g the C o m p t o n scattering to occur at the center of the G e detector, the absorption ranges x ( 0 ) in N a I a n d B G O shieldings have been calculated [eq. (3)] for different percentage values of A / A o and using incident energies ranging from 600 to 1500 keV. Fig. 2b shows a representation for the c o n t o u r of the ranges in b o t h B G O a n d N a I for 1 MeV 7-ray energy a n d assuming a 90% fractional absorption A / A o. For the same energy region, i.e. 6 0 0 - 1 5 0 0 keV, geometrical and rand o m calculations have been carried out to estimate the dimensions of the B G O a n d N a l shields as functions of A / A o as presented in fig. 3. G o o d agreement between b o t h methods of calculation can be seen for the lengths a n d diameters of the shields as well as its thickness in front of the G e detector as shown in sections a, b, c of fig. 3 respectively. Accordingly, one can a d o p t for such calculations the geometrical m e t h o d for its simplicity. Moreover, from the comparison between B G O a n d NaI(TI) curves, it can be concluded that for a fractional absorption of 90% a decrease can be reached in the
The trajectory of a g a m m a ray x(O) has been simulated [8] by the e q u a t i o n x = - (1/At) In ~/,
(4)
where ~ is a r a n d o m n u m b e r uniformly distributed on the interval from 0 to 1. The r a n d o m n u m b e r s used for the calculation have been generated by the multiplicate m e t h o d [ 11 ] ~i+,=Bi(216+3)
mod(23'-
1),
243
(5)
with 7o = 987 654 31.
5. Results and discussion The investigated spectrometer as shown in fig. 2a consists of a G e detector of dimensions ~ 5 x 5 cm located asymmetrically in a cylindrical scintillation material ( B G O or N a I ) of diameter d cm a n d length l cm. The Ge-detector is placed perpendicularly to the
SCINTILLATOR
o)
¢
7O
80
I
./
cm
60
7
oS
i
J
/
,.o/
\
.o
b)
,.o% i'to ) h~.
\
__ \ ,o /
Fig. 2. (a) Schematic cross sectional drawing of the Compton suppression spectrometer through a plane perpendicular to the axis of the Ge detector. (b) The 90% absorption range contour of Compton scattered y-rays for hv o = 1 MeV in NaI(T1) and BGO.
244
M. Mohsen / Compact Compton suppression spectrometer
BGO
lO0
o
Nol
BGO
lC
Nal
BGO
Nol
8O
Z 60 0 I,.O. n," 0 O3 ¢1 .~ 4 0 ...,I
• GEOMETRI CAL
Z 0 IU <[ (Z: la.
X RANDOM 2o
t 4
8
12
16
20
24
28
£ cm
36
4
SCINTILATOR LENGTH [o)
8
12
16
20
24
d cm 32
SCINTILLATOR DIAMETER (b)
I
2
3
4
t cm
6
DISTANCE BETWEEN OE & SCINTILLATOR FRONT (c1
Fig. 3. Geometrical and random calculations for the dimensions of BGO and NaI shields as functions of the fractional absorption A / A o for an energy of 600-1500 keV.
d i m e n s i o n s o f the B G O C o m p t o n s u p p r e s s i o n spect r o m e t e r to nearly half those of NaI. It is o f interest to a p p l y this result in p r e d i c t i n g the s h a p e o f a s p e c t r u m after s u p p r e s s i o n with BGO. O n the basis of the 6 ° C o s p e c t r u m m e a s u r e d b y A a r t s et al. [3] b e f o r e a n d after s u p p r e s s i o n with a N a I crystal o f ~ 30 c m x 35 c m as s h o w n in fig. 4, o n e can achieve a similar s u p p r e s s i o n w h e n using B G O scintillator o f ~ 14 c m x 16 cm.
T h e a u t h o r wishes to t h a n k Prof. Dr. F. El Bedewi at the K i n g A b d E1 Aziz U n i v e r s i t y a n d Dr. T.L. K h o o at A r g o n n e N a t i o n a l L a b o r a t o r y for their interest a n d stimulating discussions.
References
1173
1332
6 1o
60 CO (D I'Z
f')
~
WITHOUT S
~
~0 4
103
I
soo
~
I
coo
~ooo
Ey
I
~2oo
~4oo
(KeV)
Fig. 4. Spectrum of 6°Co as measured by Aarts et al. [3], down to an energy of 600 keV with NaI suppressor of ~ 30 cm × 35 cm which can be similar to that of BGO suppressor of ~ 14 c m × 16 cm.
[1] D.C. Camp, Proc. Int. Conf. on Radioactivity in nuclear spectroscopy, vol. 1 (Vanderbih University, August, 1969) p. 135. [2] J. Konijn, P.F.A. Goudsmit and E.W.A. Lingeman, Nucl. Instr. and Meth. 109 (1973) 83. [3] H.J.M. Aarts, C.J. Van der Poel, D.E.C. Scherpenzeel, H.F.R. Arciszewski and G.A.P. Engelbertink, private communication (1981). [4] R. Beetz, W.L. Posthumus, F.W.N. De Boer, J.L. Maarleveld, A. van der Schaaf and J. Konijn, Nuel. Instr. and Meth. 145 (1977) 353. [5] M. Mohsen, Bulletin of Faculty of Science K.A.U., Jeddah (in press). [6] A.E. Evans, Jr., IEEE Trans. Nucl. Sci. NS-27 (1980) p. 172. [7] M.J. Weber and R.R, Monchamp, J. Appl. Phys. 44 (1973) 5496. [8] R. Nitsche, J. Appl. Phys. 36 (1965) 2358. [9] D.P. Bortfeld and H. Meier, J. Appl. Phys. 43 (1972) 5110. [10] O.H. Nestor and C.Y. Huang, IEEE Trans. Nucl. Sci. NS-22 (1975) 68. [11] B.F. Peterman, S. Hontzeas and R.G. Rystephanick, Nucl. Instr. and Meth. 104 (1972) 461.
241
A COMPACT C O M P T O N S U P P R E S S I O N S P E C T R O M E T E R U S I N G B I S M U T H GERMANATE SCINTILLATOR Mona MOHSEN Department of Physics, Section of Girls, King Abd EI Aziz University, Jeddah, P.O. Box 9030, Saudi Arabia
Received 31 March 1982 and in revised from 21 December 1982
Geometrical and random sampling methods of calculation have been used to estimate the dimensions of a Compton suppression spectrometer using bismuth germanate (BGO) scintillator. Similar calculations have been done using sodium iodide. It has been found that the dimensions of the BGO shield can be decreased to nearly half those of NaI(TI).
1. Introduction In -/-ray spectroscopy dealing with reactions or decay, the complexity of the spectra and the low yield of many transitions is difficult to analyze without increasing the peak-to-background ratio. One of the ways to enhance weak transitions with energies up to 4 MeV is in the suppression of the Compton continuum. Camp [1] designed a symmetrical Ge(Li)-NaI(T1) Compton suppression spectrometer, which consists of a cylindrical scintillator shield with a coaxial hollow core for a central detector at one side and the incoming y-ray beam at the opposite side. Such a set-up suffers from the presence of collinear holes which leads to a low suppression of double escape peaks as well as Compton scattering in the low and high energy parts of the spectrum. This disadvantage could be avoided in an asymmetrical geometry, where the central detector is oriented at right angles to the direction of the incoming radiation. Using this geometry, Ge-NaI(T1) Compton suppression spectrometers have been successfully constructed and their performance has been studied in radioactive decay [2], 7 - ' / c o i n c i d e n c e [3] and in experiments dealing with in-beam '/-ray spectroscopy at the external beam of a synchrocyclotron [4]. However, it will be of value especially for multiparameter measurements to reduce the size of such a spectrometer without affecting its function and this could be achieved by replacing NaI(T1) by bismuth germanate (BiaGe3012 known as BGO), which has recently proved to be a more efficient scintillator
[5]. Table 1 shows a comparison of the optical and mechanical properties of NaI(T1) and B G O scintillators. Although the energy and time resolution of NaI is better than BGO, yet the latter is characterized by high atomic number and density, which leads to an increase 0167-5087/83/0000-0000/$03.00 © 1983 North-Holland
Table 1 Comparison of NaI(T1) and BGO scintillators. Nal(TI) Energy resolution at 137Cs-photopeak (%) 7 Light output (relative) 100 Wavelength of maximum scintillation intensity(rim) 420 Refractive index at maximum intensity 1.85 Radiation length (cm) > 1.05 Decay constant (ns) 250 After glow (relative) 100 Density (g/cm 3) 3.67 Hygroscopicity hygroscopic
BGO 15 8-12 480 2.1 1.05 300 1 7.13 non-hygroscopic
in the stopping power and the photopeak efficiency of the BGO shield. Moreover, its full energy peak efficiency at 1.33 MeV has been measured by Evans [6] and found to be greater than that of NaI(TI) by a factor of 4.5. It should also be pointed out that B G O is free of problems associated with nonuniform dopant distribution. Weber et al. [7] identified the light emitted from B G O crystals with the 3P 1 ~ 1 S 0 electronic transition within the Bi3++ ions, giving rise to a " p u r e " scintillator. Moreover, Nitsche [8], Bortfeld et al. [9] and Nestor et al. [10] measured the light transmission in Bi4Ge3Ot2 for 0.5 mm, 0.65 mm and 17,5 mm sample thicknesses respectively as functions of light wavelength. Fig. 1 shows the light transmission in BGO taken from these references as a function of sample thickness at 480 nm, which represents the wavelength of maximum scintillation intensity [10]. Even for a sample thickness of 100 cm a light transmission greater than 65% can be obtained. Accordingly, these properties can then nominate
M. Mohsen / Compact Compton suppression spectrometer
242
I
-/.
i
80
i
Z
o_ 70
(n :E 01 Z
.< rr I-
.J
60
50
0 m
5
0.I
2
I
5
I0
2
Leng'rh
I00
5 fftnl
I000
Fig. 1. BGO light transmission as a function of sample length at 480 nm wavelength.
B G O for construction of a compact Compton suppression spectrometer, whose function depends on efficiency rather than on energy and time resolution. The aim of the present work is to present a comparative study for an asymmetrical Compton suppression spectrometer using G e - B G O as well as Ge-NaI(T1). For this purpose geometrical and random sampling methods have been applied to calculate the dimensions of each spectrometer as a function of its relative absorption.
2. Principle A Compton suppression spectrometer consists of a central Ge detector surrounded by a scintillation shield (BGO or NaI). G a m m a rays from a source outside are collimated before striking the Ge detector positioned at right angle to the direction of the beam. Compton scattered gamma rays from that detector can then interact with the surrounding scintillating material. If the resulting signal is above the noise level, then it can be used to reject the recording of any coincident signal in the Ge detector. Accordingly, full energy events only should in principle remain in the spectrum of the Ge detector. However, the Compton continuum is not completely eliminated since backscattered gamma rays can escape from the scintillation crystal through the entrance hole of the incident beam. Such gamma rays leave most of their energy in the Ge detector and this contribution appears at the Compton edge of the spectrum.
3. Geometrical method of calculation In order to estimate the dimensions of a Compton suppression spectrometer geometrically, gamma events in the germanium due to single Compton collisions are assumed. The continuum spectrum without shielding can be represented as do dE
do dO dR d E '
(1)
where o is the cross section for Compton collision, E is the kinetic energy of the recoil electron in the Ge detector and R is the angle of the scattered gamma relative to the incident gamma. The quantity d o / d R is the cross section for photons scattered between two cones at 8 and R + dR. The quantity d R ~ d E is obtained from E=h%[l-(l+c~(1-cosO)}
1],
(2)
where h u0 is the energy of incident gamma rays and c~ = h 1,o/mc 2.
To determine the expected effect of the anticoincidence shield in suppressing the continuum, the absorption of the scintillator around the Ge detector is investigated according to the scattering angle 0. The fractional absorption A / A o of the scattered gammas in the appropriate path length x ( 0 ) of the scintillator is calculated from A / A o = 1 - e -~x(°),
where # is the total linear absorption coefficient of the scintillator for the gamma energy ( h % - E) at angle 0.
M. Mohsen / Compact Compton suppression spectrometer
4. Random sampling of path length
direction of the g a m m a ray incident through a hole d i a m e t e r of 1 cm at a distance t cm from the detector. A s s u m i n g the C o m p t o n scattering to occur at the center of the G e detector, the absorption ranges x ( 0 ) in N a I a n d B G O shieldings have been calculated [eq. (3)] for different percentage values of A / A o and using incident energies ranging from 600 to 1500 keV. Fig. 2b shows a representation for the c o n t o u r of the ranges in b o t h B G O a n d N a I for 1 MeV 7-ray energy a n d assuming a 90% fractional absorption A / A o. For the same energy region, i.e. 6 0 0 - 1 5 0 0 keV, geometrical and rand o m calculations have been carried out to estimate the dimensions of the B G O a n d N a l shields as functions of A / A o as presented in fig. 3. G o o d agreement between b o t h methods of calculation can be seen for the lengths a n d diameters of the shields as well as its thickness in front of the G e detector as shown in sections a, b, c of fig. 3 respectively. Accordingly, one can a d o p t for such calculations the geometrical m e t h o d for its simplicity. Moreover, from the comparison between B G O a n d NaI(TI) curves, it can be concluded that for a fractional absorption of 90% a decrease can be reached in the
The trajectory of a g a m m a ray x(O) has been simulated [8] by the e q u a t i o n x = - (1/At) In ~/,
(4)
where ~ is a r a n d o m n u m b e r uniformly distributed on the interval from 0 to 1. The r a n d o m n u m b e r s used for the calculation have been generated by the multiplicate m e t h o d [ 11 ] ~i+,=Bi(216+3)
mod(23'-
1),
243
(5)
with 7o = 987 654 31.
5. Results and discussion The investigated spectrometer as shown in fig. 2a consists of a G e detector of dimensions ~ 5 x 5 cm located asymmetrically in a cylindrical scintillation material ( B G O or N a I ) of diameter d cm a n d length l cm. The Ge-detector is placed perpendicularly to the
SCINTILLATOR
o)
¢
7O
80
I
./
cm
60
7
oS
i
J
/
,.o/
\
.o
b)
,.o% i'to ) h~.
\
__ \ ,o /
Fig. 2. (a) Schematic cross sectional drawing of the Compton suppression spectrometer through a plane perpendicular to the axis of the Ge detector. (b) The 90% absorption range contour of Compton scattered y-rays for hv o = 1 MeV in NaI(T1) and BGO.
244
M. Mohsen / Compact Compton suppression spectrometer
BGO
lO0
o
Nol
BGO
lC
Nal
BGO
Nol
8O
Z 60 0 I,.O. n," 0 O3 ¢1 .~ 4 0 ...,I
• GEOMETRI CAL
Z 0 IU <[ (Z: la.
X RANDOM 2o
t 4
8
12
16
20
24
28
£ cm
36
4
SCINTILATOR LENGTH [o)
8
12
16
20
24
d cm 32
SCINTILLATOR DIAMETER (b)
I
2
3
4
t cm
6
DISTANCE BETWEEN OE & SCINTILLATOR FRONT (c1
Fig. 3. Geometrical and random calculations for the dimensions of BGO and NaI shields as functions of the fractional absorption A / A o for an energy of 600-1500 keV.
d i m e n s i o n s o f the B G O C o m p t o n s u p p r e s s i o n spect r o m e t e r to nearly half those of NaI. It is o f interest to a p p l y this result in p r e d i c t i n g the s h a p e o f a s p e c t r u m after s u p p r e s s i o n with BGO. O n the basis of the 6 ° C o s p e c t r u m m e a s u r e d b y A a r t s et al. [3] b e f o r e a n d after s u p p r e s s i o n with a N a I crystal o f ~ 30 c m x 35 c m as s h o w n in fig. 4, o n e can achieve a similar s u p p r e s s i o n w h e n using B G O scintillator o f ~ 14 c m x 16 cm.
T h e a u t h o r wishes to t h a n k Prof. Dr. F. El Bedewi at the K i n g A b d E1 Aziz U n i v e r s i t y a n d Dr. T.L. K h o o at A r g o n n e N a t i o n a l L a b o r a t o r y for their interest a n d stimulating discussions.
References
1173
1332
6 1o
60 CO (D I'Z
f')
~
WITHOUT S
~
~0 4
103
I
soo
~
I
coo
~ooo
Ey
I
~2oo
~4oo
(KeV)
Fig. 4. Spectrum of 6°Co as measured by Aarts et al. [3], down to an energy of 600 keV with NaI suppressor of ~ 30 cm × 35 cm which can be similar to that of BGO suppressor of ~ 14 c m × 16 cm.
[1] D.C. Camp, Proc. Int. Conf. on Radioactivity in nuclear spectroscopy, vol. 1 (Vanderbih University, August, 1969) p. 135. [2] J. Konijn, P.F.A. Goudsmit and E.W.A. Lingeman, Nucl. Instr. and Meth. 109 (1973) 83. [3] H.J.M. Aarts, C.J. Van der Poel, D.E.C. Scherpenzeel, H.F.R. Arciszewski and G.A.P. Engelbertink, private communication (1981). [4] R. Beetz, W.L. Posthumus, F.W.N. De Boer, J.L. Maarleveld, A. van der Schaaf and J. Konijn, Nuel. Instr. and Meth. 145 (1977) 353. [5] M. Mohsen, Bulletin of Faculty of Science K.A.U., Jeddah (in press). [6] A.E. Evans, Jr., IEEE Trans. Nucl. Sci. NS-27 (1980) p. 172. [7] M.J. Weber and R.R, Monchamp, J. Appl. Phys. 44 (1973) 5496. [8] R. Nitsche, J. Appl. Phys. 36 (1965) 2358. [9] D.P. Bortfeld and H. Meier, J. Appl. Phys. 43 (1972) 5110. [10] O.H. Nestor and C.Y. Huang, IEEE Trans. Nucl. Sci. NS-22 (1975) 68. [11] B.F. Peterman, S. Hontzeas and R.G. Rystephanick, Nucl. Instr. and Meth. 104 (1972) 461.